diff --git a/.github/workflows/lint.yaml b/.github/workflows/lint.yaml
index e8623f86c..2e81b7ca7 100644
--- a/.github/workflows/lint.yaml
+++ b/.github/workflows/lint.yaml
@@ -19,7 +19,7 @@ jobs:
python-version: 3.8
- name: Install dependencies
- run: pip install black nbdev pre-commit
+ run: pip install black nbdev==2.3.25 pre-commit
- name: Run pre-commit
run: pre-commit run --files neuralforecast/*
diff --git a/nbs/models.timemixer.ipynb b/nbs/models.timemixer.ipynb
index f06b214a4..87ed045a5 100644
--- a/nbs/models.timemixer.ipynb
+++ b/nbs/models.timemixer.ipynb
@@ -36,8 +36,6 @@
"source": [
"#| export\n",
"\n",
- "import numpy as np\n",
- "\n",
"import torch\n",
"import torch.nn as nn\n",
"\n",
@@ -457,7 +455,7 @@
"\n",
" # Class attributes\n",
" SAMPLING_TYPE = 'multivariate'\n",
- " EXOGENOUS_FUTR = False\n",
+ " EXOGENOUS_FUTR = True\n",
" EXOGENOUS_HIST = False\n",
" EXOGENOUS_STAT = False\n",
"\n",
@@ -480,7 +478,6 @@
" down_sampling_window: int = 2,\n",
" down_sampling_method: str = 'avg',\n",
" use_norm: bool = True,\n",
- " decoder_input_size_multiplier: float = 0.5,\n",
" loss = MAE(),\n",
" valid_loss = None,\n",
" max_steps: int = 1000,\n",
@@ -525,10 +522,6 @@
" lr_scheduler_kwargs=lr_scheduler_kwargs,\n",
" **trainer_kwargs)\n",
" \n",
- " self.label_len = int(np.ceil(input_size * decoder_input_size_multiplier))\n",
- " if (self.label_len >= input_size) or (self.label_len <= 0):\n",
- " raise Exception(f'Check decoder_input_size_multiplier={decoder_input_size_multiplier}, range (0,1)')\n",
- " \n",
" self.h = h\n",
" self.input_size = input_size\n",
" self.e_layers = e_layers\n",
@@ -674,9 +667,9 @@
" if self.channel_independence == 1:\n",
" B, T, N = x_enc.size()\n",
" x_mark_dec = x_mark_dec.repeat(N, 1, 1)\n",
- " self.x_mark_dec = self.enc_embedding(None, x_mark_dec)\n",
+ " self.x_mark_dec = self.enc_embedding(None, x_mark_dec) \n",
" else:\n",
- " self.x_mark_dec = self.enc_embedding(None, x_mark_dec)\n",
+ " self.x_mark_dec = self.enc_embedding(x_mark_enc, x_mark_dec)\n",
"\n",
" x_enc, x_mark_enc = self.__multi_scale_process_inputs(x_enc, x_mark_enc)\n",
"\n",
@@ -751,8 +744,8 @@
" futr_exog = windows_batch['futr_exog']\n",
"\n",
" if self.futr_exog_size > 0:\n",
- " x_mark_enc = futr_exog[:, :, :self.input_size, :]\n",
- " x_mark_dec = futr_exog[:, :, -(self.label_len + self.h):, :]\n",
+ " x_mark_enc = futr_exog[:, :, :self.input_size]\n",
+ " x_mark_dec = None\n",
" else:\n",
" x_mark_enc = None\n",
" x_mark_dec = None\n",
@@ -773,7 +766,147 @@
"cell_type": "code",
"execution_count": null,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "---\n",
+ "\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/timemixer.py#L327){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "\n",
+ "### TimeMixer\n",
+ "\n",
+ "> TimeMixer (h, input_size, n_series, stat_exog_list=None,\n",
+ "> hist_exog_list=None, futr_exog_list=None, d_model:int=32,\n",
+ "> d_ff:int=32, dropout:float=0.1, e_layers:int=4, top_k:int=5,\n",
+ "> decomp_method:str='moving_avg', moving_avg:int=25,\n",
+ "> channel_independence:int=0, down_sampling_layers:int=1,\n",
+ "> down_sampling_window:int=2, down_sampling_method:str='avg',\n",
+ "> use_norm:bool=True, loss=MAE(), valid_loss=None,\n",
+ "> max_steps:int=1000, learning_rate:float=0.001,\n",
+ "> num_lr_decays:int=-1, early_stop_patience_steps:int=-1,\n",
+ "> val_check_steps:int=100, batch_size:int=32, step_size:int=1,\n",
+ "> scaler_type:str='identity', random_seed:int=1,\n",
+ "> num_workers_loader:int=0, drop_last_loader:bool=False,\n",
+ "> optimizer=None, optimizer_kwargs=None, lr_scheduler=None,\n",
+ "> lr_scheduler_kwargs=None, **trainer_kwargs)\n",
+ "\n",
+ "TimeMixer\n",
+ "**Parameters**
\n",
+ "`h`: int, Forecast horizon.
\n",
+ "`input_size`: int, autorregresive inputs size, y=[1,2,3,4] input_size=2 -> y_[t-2:t]=[1,2].
\n",
+ "`n_series`: int, number of time-series.
\n",
+ "`futr_exog_list`: str list, future exogenous columns.
\n",
+ "`hist_exog_list`: str list, historic exogenous columns.
\n",
+ "`stat_exog_list`: str list, static exogenous columns.
\n",
+ "`d_model`: int, dimension of the model.
\n",
+ "`d_ff`: int, dimension of the fully-connected network.
\n",
+ "`dropout`: float, dropout rate.
\n",
+ "`e_layers`: int, number of encoder layers.
\n",
+ "`top_k`: int, number of selected frequencies.
\n",
+ "`decomp_method`: str, method of series decomposition [moving_avg, dft_decomp].
\n",
+ "`moving_avg`: int, window size of moving average.
\n",
+ "`channel_independence`: int, 0: channel dependence, 1: channel independence.
\n",
+ "`down_sampling_layers`: int, number of downsampling layers.
\n",
+ "`down_sampling_window`: int, size of downsampling window.
\n",
+ "`down_sampling_method`: str, down sampling method [avg, max, conv].
\n",
+ "`use_norm`: bool, whether to normalize or not.
\n",
+ " `decoder_input_size_multiplier`: float = 0.5.
\n",
+ "`loss`: PyTorch module, instantiated train loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
+ "`valid_loss`: PyTorch module=`loss`, instantiated valid loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
+ "`max_steps`: int=1000, maximum number of training steps.
\n",
+ "`learning_rate`: float=1e-3, Learning rate between (0, 1).
\n",
+ "`num_lr_decays`: int=-1, Number of learning rate decays, evenly distributed across max_steps.
\n",
+ "`early_stop_patience_steps`: int=-1, Number of validation iterations before early stopping.
\n",
+ "`val_check_steps`: int=100, Number of training steps between every validation loss check.
\n",
+ "`batch_size`: int=32, number of different series in each batch.
\n",
+ "`step_size`: int=1, step size between each window of temporal data.
\n",
+ "`scaler_type`: str='identity', type of scaler for temporal inputs normalization see [temporal scalers](https://nixtla.github.io/neuralforecast/common.scalers.html).
\n",
+ "`random_seed`: int=1, random_seed for pytorch initializer and numpy generators.
\n",
+ "`num_workers_loader`: int=os.cpu_count(), workers to be used by `TimeSeriesDataLoader`.
\n",
+ "`drop_last_loader`: bool=False, if True `TimeSeriesDataLoader` drops last non-full batch.
\n",
+ "`alias`: str, optional, Custom name of the model.
\n",
+ "`optimizer`: Subclass of 'torch.optim.Optimizer', optional, user specified optimizer instead of the default choice (Adam).
\n",
+ "`optimizer_kwargs`: dict, optional, list of parameters used by the user specified `optimizer`.
\n",
+ "`lr_scheduler`: Subclass of 'torch.optim.lr_scheduler.LRScheduler', optional, user specified lr_scheduler instead of the default choice (StepLR).
\n",
+ "`lr_scheduler_kwargs`: dict, optional, list of parameters used by the user specified `lr_scheduler`.
\n",
+ "`**trainer_kwargs`: int, keyword trainer arguments inherited from [PyTorch Lighning's trainer](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).
\n",
+ "\n",
+ "**References**
\n",
+ "[Shiyu Wang, Haixu Wu, Xiaoming Shi, Tengge Hu, Huakun Luo, Lintao Ma, James Y. Zhang, Jun Zhou.\"TimeMixer: Decomposable Multiscale Mixing For Time Series Forecasting\"](https://openreview.net/pdf?id=7oLshfEIC2)"
+ ],
+ "text/plain": [
+ "---\n",
+ "\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/timemixer.py#L327){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "\n",
+ "### TimeMixer\n",
+ "\n",
+ "> TimeMixer (h, input_size, n_series, stat_exog_list=None,\n",
+ "> hist_exog_list=None, futr_exog_list=None, d_model:int=32,\n",
+ "> d_ff:int=32, dropout:float=0.1, e_layers:int=4, top_k:int=5,\n",
+ "> decomp_method:str='moving_avg', moving_avg:int=25,\n",
+ "> channel_independence:int=0, down_sampling_layers:int=1,\n",
+ "> down_sampling_window:int=2, down_sampling_method:str='avg',\n",
+ "> use_norm:bool=True, loss=MAE(), valid_loss=None,\n",
+ "> max_steps:int=1000, learning_rate:float=0.001,\n",
+ "> num_lr_decays:int=-1, early_stop_patience_steps:int=-1,\n",
+ "> val_check_steps:int=100, batch_size:int=32, step_size:int=1,\n",
+ "> scaler_type:str='identity', random_seed:int=1,\n",
+ "> num_workers_loader:int=0, drop_last_loader:bool=False,\n",
+ "> optimizer=None, optimizer_kwargs=None, lr_scheduler=None,\n",
+ "> lr_scheduler_kwargs=None, **trainer_kwargs)\n",
+ "\n",
+ "TimeMixer\n",
+ "**Parameters**
\n",
+ "`h`: int, Forecast horizon.
\n",
+ "`input_size`: int, autorregresive inputs size, y=[1,2,3,4] input_size=2 -> y_[t-2:t]=[1,2].
\n",
+ "`n_series`: int, number of time-series.
\n",
+ "`futr_exog_list`: str list, future exogenous columns.
\n",
+ "`hist_exog_list`: str list, historic exogenous columns.
\n",
+ "`stat_exog_list`: str list, static exogenous columns.
\n",
+ "`d_model`: int, dimension of the model.
\n",
+ "`d_ff`: int, dimension of the fully-connected network.
\n",
+ "`dropout`: float, dropout rate.
\n",
+ "`e_layers`: int, number of encoder layers.
\n",
+ "`top_k`: int, number of selected frequencies.
\n",
+ "`decomp_method`: str, method of series decomposition [moving_avg, dft_decomp].
\n",
+ "`moving_avg`: int, window size of moving average.
\n",
+ "`channel_independence`: int, 0: channel dependence, 1: channel independence.
\n",
+ "`down_sampling_layers`: int, number of downsampling layers.
\n",
+ "`down_sampling_window`: int, size of downsampling window.
\n",
+ "`down_sampling_method`: str, down sampling method [avg, max, conv].
\n",
+ "`use_norm`: bool, whether to normalize or not.
\n",
+ " `decoder_input_size_multiplier`: float = 0.5.
\n",
+ "`loss`: PyTorch module, instantiated train loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
+ "`valid_loss`: PyTorch module=`loss`, instantiated valid loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
+ "`max_steps`: int=1000, maximum number of training steps.
\n",
+ "`learning_rate`: float=1e-3, Learning rate between (0, 1).
\n",
+ "`num_lr_decays`: int=-1, Number of learning rate decays, evenly distributed across max_steps.
\n",
+ "`early_stop_patience_steps`: int=-1, Number of validation iterations before early stopping.
\n",
+ "`val_check_steps`: int=100, Number of training steps between every validation loss check.
\n",
+ "`batch_size`: int=32, number of different series in each batch.
\n",
+ "`step_size`: int=1, step size between each window of temporal data.
\n",
+ "`scaler_type`: str='identity', type of scaler for temporal inputs normalization see [temporal scalers](https://nixtla.github.io/neuralforecast/common.scalers.html).
\n",
+ "`random_seed`: int=1, random_seed for pytorch initializer and numpy generators.
\n",
+ "`num_workers_loader`: int=os.cpu_count(), workers to be used by `TimeSeriesDataLoader`.
\n",
+ "`drop_last_loader`: bool=False, if True `TimeSeriesDataLoader` drops last non-full batch.
\n",
+ "`alias`: str, optional, Custom name of the model.
\n",
+ "`optimizer`: Subclass of 'torch.optim.Optimizer', optional, user specified optimizer instead of the default choice (Adam).
\n",
+ "`optimizer_kwargs`: dict, optional, list of parameters used by the user specified `optimizer`.
\n",
+ "`lr_scheduler`: Subclass of 'torch.optim.lr_scheduler.LRScheduler', optional, user specified lr_scheduler instead of the default choice (StepLR).
\n",
+ "`lr_scheduler_kwargs`: dict, optional, list of parameters used by the user specified `lr_scheduler`.
\n",
+ "`**trainer_kwargs`: int, keyword trainer arguments inherited from [PyTorch Lighning's trainer](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).
\n",
+ "\n",
+ "**References**
\n",
+ "[Shiyu Wang, Haixu Wu, Xiaoming Shi, Tengge Hu, Huakun Luo, Lintao Ma, James Y. Zhang, Jun Zhou.\"TimeMixer: Decomposable Multiscale Mixing For Time Series Forecasting\"](https://openreview.net/pdf?id=7oLshfEIC2)"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"show_doc(TimeMixer)"
]
@@ -782,7 +915,71 @@
"cell_type": "code",
"execution_count": null,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "---\n",
+ "\n",
+ "### TimeMixer.fit\n",
+ "\n",
+ "> TimeMixer.fit (dataset, val_size=0, test_size=0, random_seed=None,\n",
+ "> distributed_config=None)\n",
+ "\n",
+ "Fit.\n",
+ "\n",
+ "The `fit` method, optimizes the neural network's weights using the\n",
+ "initialization parameters (`learning_rate`, `windows_batch_size`, ...)\n",
+ "and the `loss` function as defined during the initialization.\n",
+ "Within `fit` we use a PyTorch Lightning `Trainer` that\n",
+ "inherits the initialization's `self.trainer_kwargs`, to customize\n",
+ "its inputs, see [PL's trainer arguments](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).\n",
+ "\n",
+ "The method is designed to be compatible with SKLearn-like classes\n",
+ "and in particular to be compatible with the StatsForecast library.\n",
+ "\n",
+ "By default the `model` is not saving training checkpoints to protect\n",
+ "disk memory, to get them change `enable_checkpointing=True` in `__init__`.\n",
+ "\n",
+ "**Parameters:**
\n",
+ "`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
+ "`val_size`: int, validation size for temporal cross-validation.
\n",
+ "`test_size`: int, test size for temporal cross-validation.
"
+ ],
+ "text/plain": [
+ "---\n",
+ "\n",
+ "### TimeMixer.fit\n",
+ "\n",
+ "> TimeMixer.fit (dataset, val_size=0, test_size=0, random_seed=None,\n",
+ "> distributed_config=None)\n",
+ "\n",
+ "Fit.\n",
+ "\n",
+ "The `fit` method, optimizes the neural network's weights using the\n",
+ "initialization parameters (`learning_rate`, `windows_batch_size`, ...)\n",
+ "and the `loss` function as defined during the initialization.\n",
+ "Within `fit` we use a PyTorch Lightning `Trainer` that\n",
+ "inherits the initialization's `self.trainer_kwargs`, to customize\n",
+ "its inputs, see [PL's trainer arguments](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.trainer.trainer.Trainer.html?highlight=trainer).\n",
+ "\n",
+ "The method is designed to be compatible with SKLearn-like classes\n",
+ "and in particular to be compatible with the StatsForecast library.\n",
+ "\n",
+ "By default the `model` is not saving training checkpoints to protect\n",
+ "disk memory, to get them change `enable_checkpointing=True` in `__init__`.\n",
+ "\n",
+ "**Parameters:**
\n",
+ "`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
+ "`val_size`: int, validation size for temporal cross-validation.
\n",
+ "`test_size`: int, test size for temporal cross-validation.
"
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"show_doc(TimeMixer.fit, name='TimeMixer.fit')"
]
@@ -791,7 +988,51 @@
"cell_type": "code",
"execution_count": null,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/markdown": [
+ "---\n",
+ "\n",
+ "### TimeMixer.predict\n",
+ "\n",
+ "> TimeMixer.predict (dataset, test_size=None, step_size=1,\n",
+ "> random_seed=None, **data_module_kwargs)\n",
+ "\n",
+ "Predict.\n",
+ "\n",
+ "Neural network prediction with PL's `Trainer` execution of `predict_step`.\n",
+ "\n",
+ "**Parameters:**
\n",
+ "`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
+ "`test_size`: int=None, test size for temporal cross-validation.
\n",
+ "`step_size`: int=1, Step size between each window.
\n",
+ "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule)."
+ ],
+ "text/plain": [
+ "---\n",
+ "\n",
+ "### TimeMixer.predict\n",
+ "\n",
+ "> TimeMixer.predict (dataset, test_size=None, step_size=1,\n",
+ "> random_seed=None, **data_module_kwargs)\n",
+ "\n",
+ "Predict.\n",
+ "\n",
+ "Neural network prediction with PL's `Trainer` execution of `predict_step`.\n",
+ "\n",
+ "**Parameters:**
\n",
+ "`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
+ "`test_size`: int=None, test size for temporal cross-validation.
\n",
+ "`step_size`: int=1, Step size between each window.
\n",
+ "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule)."
+ ]
+ },
+ "execution_count": null,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"show_doc(TimeMixer.predict, name='TimeMixer.predict')"
]
@@ -816,14 +1057,395 @@
"from neuralforecast import NeuralForecast\n",
"from neuralforecast.models import TimeMixer\n",
"from neuralforecast.utils import AirPassengersPanel, AirPassengersStatic\n",
- "from neuralforecast.losses.pytorch import MAE, DistributionLoss"
+ "from neuralforecast.losses.pytorch import MAE, MSE"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Seed set to 1\n",
+ "GPU available: True (mps), used: True\n",
+ "TPU available: False, using: 0 TPU cores\n",
+ "IPU available: False, using: 0 IPUs\n",
+ "HPU available: False, using: 0 HPUs\n",
+ "\n",
+ " | Name | Type | Params\n",
+ "------------------------------------------------------------\n",
+ "0 | loss | MAE | 0 \n",
+ "1 | valid_loss | MAE | 0 \n",
+ "2 | padder | ConstantPad1d | 0 \n",
+ "3 | scaler | TemporalNorm | 0 \n",
+ "4 | pdm_blocks | ModuleList | 22.6 K\n",
+ "5 | preprocess | SeriesDecomp | 0 \n",
+ "6 | enc_embedding | DataEmbedding_wo_pos | 2.6 K \n",
+ "7 | normalize_layers | ModuleList | 8 \n",
+ "8 | predict_layers | ModuleList | 456 \n",
+ "9 | projection_layer | Linear | 66 \n",
+ "10 | out_res_layers | ModuleList | 756 \n",
+ "11 | regression_layers | ModuleList | 456 \n",
+ "------------------------------------------------------------\n",
+ "24.6 K Trainable params\n",
+ "2.4 K Non-trainable params\n",
+ "27.0 K Total params\n",
+ "0.108 Total estimated model params size (MB)\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "07f36aaaf2b74582b7416cfa60316011",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Sanity Checking: | | 0/? [00:00"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"#| eval: false\n",
"# Plot predictions\n",
@@ -881,7 +1514,379 @@
"cell_type": "code",
"execution_count": null,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "GPU available: True (mps), used: True\n",
+ "TPU available: False, using: 0 TPU cores\n",
+ "IPU available: False, using: 0 IPUs\n",
+ "HPU available: False, using: 0 HPUs\n",
+ "\n",
+ " | Name | Type | Params\n",
+ "------------------------------------------------------------\n",
+ "0 | loss | MAE | 0 \n",
+ "1 | valid_loss | MAE | 0 \n",
+ "2 | padder | ConstantPad1d | 0 \n",
+ "3 | scaler | TemporalNorm | 0 \n",
+ "4 | pdm_blocks | ModuleList | 22.6 K\n",
+ "5 | preprocess | SeriesDecomp | 0 \n",
+ "6 | enc_embedding | DataEmbedding_wo_pos | 2.6 K \n",
+ "7 | normalize_layers | ModuleList | 8 \n",
+ "8 | predict_layers | ModuleList | 456 \n",
+ "9 | projection_layer | Linear | 66 \n",
+ "10 | out_res_layers | ModuleList | 756 \n",
+ "11 | regression_layers | ModuleList | 456 \n",
+ "------------------------------------------------------------\n",
+ "24.6 K Trainable params\n",
+ "2.4 K Non-trainable params\n",
+ "27.0 K Total params\n",
+ "0.108 Total estimated model params size (MB)\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "90ed6397e68147b489f26568e03f39aa",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Sanity Checking: | | 0/? [00:00 16\u001b[0m \u001b[43mfcst\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdf\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mY_train_df\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mstatic_df\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mAirPassengersStatic\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mval_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m12\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m 17\u001b[0m forecasts \u001b[38;5;241m=\u001b[39m fcst\u001b[38;5;241m.\u001b[39mpredict(futr_df\u001b[38;5;241m=\u001b[39mY_test_df)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/core.py:543\u001b[0m, in \u001b[0;36mNeuralForecast.fit\u001b[0;34m(self, df, static_df, val_size, sort_df, use_init_models, verbose, id_col, time_col, target_col, distributed_config)\u001b[0m\n\u001b[1;32m 540\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_reset_models()\n\u001b[1;32m 542\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i, model \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodels):\n\u001b[0;32m--> 543\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodels[i] \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 544\u001b[0m \u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdataset\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mval_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mval_size\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdistributed_config\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdistributed_config\u001b[49m\n\u001b[1;32m 545\u001b[0m \u001b[43m \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 547\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fitted \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/common/_base_multivariate.py:547\u001b[0m, in \u001b[0;36mBaseMultivariate.fit\u001b[0;34m(self, dataset, val_size, test_size, random_seed, distributed_config)\u001b[0m\n\u001b[1;32m 543\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m distributed_config \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m 544\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m 545\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmultivariate models cannot be trained using distributed data parallel.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 546\u001b[0m )\n\u001b[0;32m--> 547\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_fit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 548\u001b[0m \u001b[43m \u001b[49m\u001b[43mdataset\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdataset\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 549\u001b[0m \u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mn_series\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 550\u001b[0m \u001b[43m \u001b[49m\u001b[43mvalid_batch_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mn_series\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 551\u001b[0m \u001b[43m \u001b[49m\u001b[43mval_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mval_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 552\u001b[0m \u001b[43m \u001b[49m\u001b[43mtest_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtest_size\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 553\u001b[0m \u001b[43m \u001b[49m\u001b[43mrandom_seed\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrandom_seed\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 554\u001b[0m \u001b[43m \u001b[49m\u001b[43mshuffle_train\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m 555\u001b[0m \u001b[43m \u001b[49m\u001b[43mdistributed_config\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m 556\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/common/_base_model.py:356\u001b[0m, in \u001b[0;36mBaseModel._fit\u001b[0;34m(self, dataset, batch_size, valid_batch_size, val_size, test_size, random_seed, shuffle_train, distributed_config)\u001b[0m\n\u001b[1;32m 354\u001b[0m model \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\n\u001b[1;32m 355\u001b[0m trainer \u001b[38;5;241m=\u001b[39m pl\u001b[38;5;241m.\u001b[39mTrainer(\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mmodel\u001b[38;5;241m.\u001b[39mtrainer_kwargs)\n\u001b[0;32m--> 356\u001b[0m \u001b[43mtrainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdatamodule\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdatamodule\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 357\u001b[0m model\u001b[38;5;241m.\u001b[39mmetrics \u001b[38;5;241m=\u001b[39m trainer\u001b[38;5;241m.\u001b[39mcallback_metrics\n\u001b[1;32m 358\u001b[0m model\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__dict__\u001b[39m\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_trainer\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:544\u001b[0m, in \u001b[0;36mTrainer.fit\u001b[0;34m(self, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path)\u001b[0m\n\u001b[1;32m 542\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mstatus \u001b[38;5;241m=\u001b[39m TrainerStatus\u001b[38;5;241m.\u001b[39mRUNNING\n\u001b[1;32m 543\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[0;32m--> 544\u001b[0m \u001b[43mcall\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_and_handle_interrupt\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 545\u001b[0m \u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_fit_impl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrain_dataloaders\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mval_dataloaders\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdatamodule\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mckpt_path\u001b[49m\n\u001b[1;32m 546\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py:44\u001b[0m, in \u001b[0;36m_call_and_handle_interrupt\u001b[0;34m(trainer, trainer_fn, *args, **kwargs)\u001b[0m\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m trainer\u001b[38;5;241m.\u001b[39mstrategy\u001b[38;5;241m.\u001b[39mlauncher \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m 43\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m trainer\u001b[38;5;241m.\u001b[39mstrategy\u001b[38;5;241m.\u001b[39mlauncher\u001b[38;5;241m.\u001b[39mlaunch(trainer_fn, \u001b[38;5;241m*\u001b[39margs, trainer\u001b[38;5;241m=\u001b[39mtrainer, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m---> 44\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtrainer_fn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 46\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m _TunerExitException:\n\u001b[1;32m 47\u001b[0m _call_teardown_hook(trainer)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:580\u001b[0m, in \u001b[0;36mTrainer._fit_impl\u001b[0;34m(self, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path)\u001b[0m\n\u001b[1;32m 573\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mfn \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m 574\u001b[0m ckpt_path \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_checkpoint_connector\u001b[38;5;241m.\u001b[39m_select_ckpt_path(\n\u001b[1;32m 575\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mfn,\n\u001b[1;32m 576\u001b[0m ckpt_path,\n\u001b[1;32m 577\u001b[0m model_provided\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m,\n\u001b[1;32m 578\u001b[0m model_connected\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlightning_module \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[1;32m 579\u001b[0m )\n\u001b[0;32m--> 580\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mckpt_path\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mckpt_path\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 582\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mstopped\n\u001b[1;32m 583\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:989\u001b[0m, in \u001b[0;36mTrainer._run\u001b[0;34m(self, model, ckpt_path)\u001b[0m\n\u001b[1;32m 984\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_signal_connector\u001b[38;5;241m.\u001b[39mregister_signal_handlers()\n\u001b[1;32m 986\u001b[0m \u001b[38;5;66;03m# ----------------------------\u001b[39;00m\n\u001b[1;32m 987\u001b[0m \u001b[38;5;66;03m# RUN THE TRAINER\u001b[39;00m\n\u001b[1;32m 988\u001b[0m \u001b[38;5;66;03m# ----------------------------\u001b[39;00m\n\u001b[0;32m--> 989\u001b[0m results \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run_stage\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 991\u001b[0m \u001b[38;5;66;03m# ----------------------------\u001b[39;00m\n\u001b[1;32m 992\u001b[0m \u001b[38;5;66;03m# POST-Training CLEAN UP\u001b[39;00m\n\u001b[1;32m 993\u001b[0m \u001b[38;5;66;03m# ----------------------------\u001b[39;00m\n\u001b[1;32m 994\u001b[0m log\u001b[38;5;241m.\u001b[39mdebug(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m: trainer tearing down\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:1033\u001b[0m, in \u001b[0;36mTrainer._run_stage\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1031\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining:\n\u001b[1;32m 1032\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m isolate_rng():\n\u001b[0;32m-> 1033\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run_sanity_check\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 1034\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m torch\u001b[38;5;241m.\u001b[39mautograd\u001b[38;5;241m.\u001b[39mset_detect_anomaly(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_detect_anomaly):\n\u001b[1;32m 1035\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfit_loop\u001b[38;5;241m.\u001b[39mrun()\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/trainer.py:1062\u001b[0m, in \u001b[0;36mTrainer._run_sanity_check\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1059\u001b[0m call\u001b[38;5;241m.\u001b[39m_call_callback_hooks(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mon_sanity_check_start\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 1061\u001b[0m \u001b[38;5;66;03m# run eval step\u001b[39;00m\n\u001b[0;32m-> 1062\u001b[0m \u001b[43mval_loop\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 1064\u001b[0m call\u001b[38;5;241m.\u001b[39m_call_callback_hooks(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mon_sanity_check_end\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 1066\u001b[0m \u001b[38;5;66;03m# reset logger connector\u001b[39;00m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/loops/utilities.py:182\u001b[0m, in \u001b[0;36m_no_grad_context.._decorator\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 180\u001b[0m context_manager \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mno_grad\n\u001b[1;32m 181\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m context_manager():\n\u001b[0;32m--> 182\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mloop_run\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/loops/evaluation_loop.py:134\u001b[0m, in \u001b[0;36m_EvaluationLoop.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 132\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbatch_progress\u001b[38;5;241m.\u001b[39mis_last_batch \u001b[38;5;241m=\u001b[39m data_fetcher\u001b[38;5;241m.\u001b[39mdone\n\u001b[1;32m 133\u001b[0m \u001b[38;5;66;03m# run step hooks\u001b[39;00m\n\u001b[0;32m--> 134\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_evaluation_step\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbatch\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbatch_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdataloader_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdataloader_iter\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 135\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mStopIteration\u001b[39;00m:\n\u001b[1;32m 136\u001b[0m \u001b[38;5;66;03m# this needs to wrap the `*_step` call too (not just `next`) for `dataloader_iter` support\u001b[39;00m\n\u001b[1;32m 137\u001b[0m \u001b[38;5;28;01mbreak\u001b[39;00m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/loops/evaluation_loop.py:391\u001b[0m, in \u001b[0;36m_EvaluationLoop._evaluation_step\u001b[0;34m(self, batch, batch_idx, dataloader_idx, dataloader_iter)\u001b[0m\n\u001b[1;32m 385\u001b[0m hook_name \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtest_step\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m trainer\u001b[38;5;241m.\u001b[39mtesting \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mvalidation_step\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 386\u001b[0m step_args \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m 387\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_build_step_args_from_hook_kwargs(hook_kwargs, hook_name)\n\u001b[1;32m 388\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m using_dataloader_iter\n\u001b[1;32m 389\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m (dataloader_iter,)\n\u001b[1;32m 390\u001b[0m )\n\u001b[0;32m--> 391\u001b[0m output \u001b[38;5;241m=\u001b[39m \u001b[43mcall\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_strategy_hook\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtrainer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mhook_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mstep_args\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 393\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbatch_progress\u001b[38;5;241m.\u001b[39mincrement_processed()\n\u001b[1;32m 395\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m using_dataloader_iter:\n\u001b[1;32m 396\u001b[0m \u001b[38;5;66;03m# update the hook kwargs now that the step method might have consumed the iterator\u001b[39;00m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/trainer/call.py:309\u001b[0m, in \u001b[0;36m_call_strategy_hook\u001b[0;34m(trainer, hook_name, *args, **kwargs)\u001b[0m\n\u001b[1;32m 306\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m 308\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m trainer\u001b[38;5;241m.\u001b[39mprofiler\u001b[38;5;241m.\u001b[39mprofile(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m[Strategy]\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mtrainer\u001b[38;5;241m.\u001b[39mstrategy\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mhook_name\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m):\n\u001b[0;32m--> 309\u001b[0m output \u001b[38;5;241m=\u001b[39m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 311\u001b[0m \u001b[38;5;66;03m# restore current_fx when nested context\u001b[39;00m\n\u001b[1;32m 312\u001b[0m pl_module\u001b[38;5;241m.\u001b[39m_current_fx_name \u001b[38;5;241m=\u001b[39m prev_fx_name\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/pytorch_lightning/strategies/strategy.py:403\u001b[0m, in \u001b[0;36mStrategy.validation_step\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 401\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlightning_module:\n\u001b[1;32m 402\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_redirection(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlightning_module, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mvalidation_step\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m--> 403\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlightning_module\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mvalidation_step\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/common/_base_multivariate.py:429\u001b[0m, in \u001b[0;36mBaseMultivariate.validation_step\u001b[0;34m(self, batch, batch_idx)\u001b[0m\n\u001b[1;32m 420\u001b[0m windows_batch \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mdict\u001b[39m(\n\u001b[1;32m 421\u001b[0m insample_y\u001b[38;5;241m=\u001b[39minsample_y, \u001b[38;5;66;03m# [Ws, L, n_series]\u001b[39;00m\n\u001b[1;32m 422\u001b[0m insample_mask\u001b[38;5;241m=\u001b[39minsample_mask, \u001b[38;5;66;03m# [Ws, L, n_series]\u001b[39;00m\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 425\u001b[0m stat_exog\u001b[38;5;241m=\u001b[39mstat_exog,\n\u001b[1;32m 426\u001b[0m ) \u001b[38;5;66;03m# [n_series, S]\u001b[39;00m\n\u001b[1;32m 428\u001b[0m \u001b[38;5;66;03m# Model Predictions\u001b[39;00m\n\u001b[0;32m--> 429\u001b[0m output \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mwindows_batch\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 430\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mloss\u001b[38;5;241m.\u001b[39mis_distribution_output:\n\u001b[1;32m 431\u001b[0m outsample_y, y_loc, y_scale \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_inv_normalization(\n\u001b[1;32m 432\u001b[0m y_hat\u001b[38;5;241m=\u001b[39moutsample_y, temporal_cols\u001b[38;5;241m=\u001b[39mbatch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtemporal_cols\u001b[39m\u001b[38;5;124m\"\u001b[39m], y_idx\u001b[38;5;241m=\u001b[39my_idx\n\u001b[1;32m 433\u001b[0m )\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m 1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m 1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m 1499\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m 1500\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m 1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/models/timemixer.py:715\u001b[0m, in \u001b[0;36mTimeMixer.forward\u001b[0;34m(self, windows_batch)\u001b[0m\n\u001b[1;32m 712\u001b[0m x_mark_enc \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m 713\u001b[0m x_mark_dec \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m--> 715\u001b[0m y_pred \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mforecast\u001b[49m\u001b[43m(\u001b[49m\u001b[43minsample_y\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_mark_enc\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_mark_dec\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 716\u001b[0m y_pred \u001b[38;5;241m=\u001b[39m y_pred[:, \u001b[38;5;241m-\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mh :, :]\n\u001b[1;32m 717\u001b[0m y_pred \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mloss\u001b[38;5;241m.\u001b[39mdomain_map(y_pred)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/models/timemixer.py:623\u001b[0m, in \u001b[0;36mTimeMixer.forecast\u001b[0;34m(self, x_enc, x_mark_enc, x_mark_dec)\u001b[0m\n\u001b[1;32m 621\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mx_mark_dec \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39menc_embedding(\u001b[38;5;28;01mNone\u001b[39;00m, x_mark_dec)\n\u001b[1;32m 622\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 623\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mx_mark_dec \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43menc_embedding\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx_mark_dec\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 625\u001b[0m x_enc, x_mark_enc \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m__multi_scale_process_inputs(x_enc, x_mark_enc)\n\u001b[1;32m 627\u001b[0m x_list \u001b[38;5;241m=\u001b[39m []\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m 1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m 1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m 1499\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m 1500\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m 1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/models/timemixer.py:120\u001b[0m, in \u001b[0;36mDataEmbedding_wo_pos.forward\u001b[0;34m(self, x, x_mark)\u001b[0m\n\u001b[1;32m 118\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtemporal_embedding(x_mark)\n\u001b[1;32m 119\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m x_mark \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 120\u001b[0m x \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mvalue_embedding\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 121\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 122\u001b[0m x \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mvalue_embedding(x) \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtemporal_embedding(x_mark)\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m 1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m 1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m 1499\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m 1500\u001b[0m \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m 1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
+ "File \u001b[0;32m~/miniconda3/envs/neuralforecast/lib/python3.10/site-packages/neuralforecast/common/_modules.py:388\u001b[0m, in \u001b[0;36mTokenEmbedding.forward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m 387\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mforward\u001b[39m(\u001b[38;5;28mself\u001b[39m, x):\n\u001b[0;32m--> 388\u001b[0m x \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtokenConv(\u001b[43mx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpermute\u001b[49m(\u001b[38;5;241m0\u001b[39m, \u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m1\u001b[39m))\u001b[38;5;241m.\u001b[39mtranspose(\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m)\n\u001b[1;32m 389\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m x\n",
+ "\u001b[0;31mAttributeError\u001b[0m: 'NoneType' object has no attribute 'permute'"
+ ]
+ }
+ ],
"source": [
"#| eval: false\n",
- "# Plot predictions\n",
- "fig, ax = plt.subplots(1, 1, figsize = (20, 7))\n",
- "Y_hat_df = forecasts.loc['Airline1']\n",
- "Y_df = AirPassengersPanel[AirPassengersPanel['unique_id']=='Airline1']\n",
"\n",
- "plt.plot(Y_df['ds'], Y_df['y'], c='black', label='True')\n",
- "plt.plot(Y_hat_df['ds'], Y_hat_df['TimeMixer'], c='blue', label='Forecast')\n",
- "ax.set_title('AirPassengers Forecast', fontsize=22)\n",
- "ax.set_ylabel('Monthly Passengers', fontsize=20)\n",
- "ax.set_xlabel('Year', fontsize=20)\n",
- "ax.legend(prop={'size': 15})\n",
- "ax.grid()"
+ "Y_train_df = AirPassengersPanel[AirPassengersPanel.ds=AirPassengersPanel['ds'].values[-12]].reset_index(drop=True) # 12 test\n",
+ "\n",
+ "model = TimeMixer(h=12,\n",
+ " input_size=24,\n",
+ " n_series=2,\n",
+ " futr_exog_list=['trend'],\n",
+ " loss=MSE(),\n",
+ " valid_loss=MAE(),\n",
+ " early_stop_patience_steps=3,\n",
+ " batch_size=32)\n",
+ "\n",
+ "fcst = NeuralForecast(models=[model], freq='M')\n",
+ "fcst.fit(df=Y_train_df, static_df=AirPassengersStatic, val_size=12)\n",
+ "forecasts = fcst.predict(futr_df=Y_test_df)"
]
}
],
diff --git a/neuralforecast/_modidx.py b/neuralforecast/_modidx.py
index 4b4d5a080..783a82704 100644
--- a/neuralforecast/_modidx.py
+++ b/neuralforecast/_modidx.py
@@ -500,7 +500,60 @@
'neuralforecast/losses/pytorch.py'),
'neuralforecast.losses.pytorch.weighted_average': ( 'losses.pytorch.html#weighted_average',
'neuralforecast/losses/pytorch.py')},
- 'neuralforecast.models.autoformer': {},
+ 'neuralforecast.models.autoformer': { 'neuralforecast.models.autoformer.AutoCorrelation': ( 'models.autoformer.html#autocorrelation',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelation.__init__': ( 'models.autoformer.html#autocorrelation.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelation.forward': ( 'models.autoformer.html#autocorrelation.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelation.time_delay_agg_full': ( 'models.autoformer.html#autocorrelation.time_delay_agg_full',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelation.time_delay_agg_inference': ( 'models.autoformer.html#autocorrelation.time_delay_agg_inference',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelation.time_delay_agg_training': ( 'models.autoformer.html#autocorrelation.time_delay_agg_training',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelationLayer': ( 'models.autoformer.html#autocorrelationlayer',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelationLayer.__init__': ( 'models.autoformer.html#autocorrelationlayer.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.AutoCorrelationLayer.forward': ( 'models.autoformer.html#autocorrelationlayer.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Autoformer': ( 'models.autoformer.html#autoformer',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Autoformer.__init__': ( 'models.autoformer.html#autoformer.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Autoformer.forward': ( 'models.autoformer.html#autoformer.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Decoder': ( 'models.autoformer.html#decoder',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Decoder.__init__': ( 'models.autoformer.html#decoder.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Decoder.forward': ( 'models.autoformer.html#decoder.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.DecoderLayer': ( 'models.autoformer.html#decoderlayer',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.DecoderLayer.__init__': ( 'models.autoformer.html#decoderlayer.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.DecoderLayer.forward': ( 'models.autoformer.html#decoderlayer.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Encoder': ( 'models.autoformer.html#encoder',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Encoder.__init__': ( 'models.autoformer.html#encoder.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.Encoder.forward': ( 'models.autoformer.html#encoder.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.EncoderLayer': ( 'models.autoformer.html#encoderlayer',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.EncoderLayer.__init__': ( 'models.autoformer.html#encoderlayer.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.EncoderLayer.forward': ( 'models.autoformer.html#encoderlayer.forward',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.LayerNorm': ( 'models.autoformer.html#layernorm',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.LayerNorm.__init__': ( 'models.autoformer.html#layernorm.__init__',
+ 'neuralforecast/models/autoformer.py'),
+ 'neuralforecast.models.autoformer.LayerNorm.forward': ( 'models.autoformer.html#layernorm.forward',
+ 'neuralforecast/models/autoformer.py')},
'neuralforecast.models.bitcn': { 'neuralforecast.models.bitcn.BiTCN': ( 'models.bitcn.html#bitcn',
'neuralforecast/models/bitcn.py'),
'neuralforecast.models.bitcn.BiTCN.__init__': ( 'models.bitcn.html#bitcn.__init__',
diff --git a/neuralforecast/models/timemixer.py b/neuralforecast/models/timemixer.py
index 755fdfd0d..ed9615a90 100644
--- a/neuralforecast/models/timemixer.py
+++ b/neuralforecast/models/timemixer.py
@@ -5,8 +5,6 @@
'PastDecomposableMixing', 'TimeMixer']
# %% ../../nbs/models.timemixer.ipynb 3
-import numpy as np
-
import torch
import torch.nn as nn
@@ -374,7 +372,7 @@ class TimeMixer(BaseMultivariate):
# Class attributes
SAMPLING_TYPE = "multivariate"
- EXOGENOUS_FUTR = False
+ EXOGENOUS_FUTR = True
EXOGENOUS_HIST = False
EXOGENOUS_STAT = False
@@ -398,7 +396,6 @@ def __init__(
down_sampling_window: int = 2,
down_sampling_method: str = "avg",
use_norm: bool = True,
- decoder_input_size_multiplier: float = 0.5,
loss=MAE(),
valid_loss=None,
max_steps: int = 1000,
@@ -416,7 +413,7 @@ def __init__(
optimizer_kwargs=None,
lr_scheduler=None,
lr_scheduler_kwargs=None,
- **trainer_kwargs,
+ **trainer_kwargs
):
super(TimeMixer, self).__init__(
@@ -443,15 +440,9 @@ def __init__(
optimizer_kwargs=optimizer_kwargs,
lr_scheduler=lr_scheduler,
lr_scheduler_kwargs=lr_scheduler_kwargs,
- **trainer_kwargs,
+ **trainer_kwargs
)
- self.label_len = int(np.ceil(input_size * decoder_input_size_multiplier))
- if (self.label_len >= input_size) or (self.label_len <= 0):
- raise Exception(
- f"Check decoder_input_size_multiplier={decoder_input_size_multiplier}, range (0,1)"
- )
-
self.h = h
self.input_size = input_size
self.e_layers = e_layers
@@ -629,7 +620,7 @@ def forecast(self, x_enc, x_mark_enc, x_mark_dec):
x_mark_dec = x_mark_dec.repeat(N, 1, 1)
self.x_mark_dec = self.enc_embedding(None, x_mark_dec)
else:
- self.x_mark_dec = self.enc_embedding(None, x_mark_dec)
+ self.x_mark_dec = self.enc_embedding(x_mark_enc, x_mark_dec)
x_enc, x_mark_enc = self.__multi_scale_process_inputs(x_enc, x_mark_enc)
@@ -715,8 +706,8 @@ def forward(self, windows_batch):
futr_exog = windows_batch["futr_exog"]
if self.futr_exog_size > 0:
- x_mark_enc = futr_exog[:, :, : self.input_size, :]
- x_mark_dec = futr_exog[:, :, -(self.label_len + self.h) :, :]
+ x_mark_enc = futr_exog[:, :, : self.input_size]
+ x_mark_dec = None
else:
x_mark_enc = None
x_mark_dec = None